An investigation of neural mechanisms of pronominal and causal inferences during reading: Evidence from NIRS

An investigation of neural mechanisms of pronominal and causal inferences during reading: Evidence from NIRS

First Author: Ms. Wanshin Chang -- National Chung Cheng University
Additional authors/chairs: 
Zih-Yun Yang; Hsin-Chin Chen; Meiyao Wu; Yuhtsuen Tzeng
Keywords: Inference, causality, pronoun, NIRS
Abstract / Summary: 

Purpose
Both pronominal and causal inferences are key mechanisms in reading comprehension, but not much research investigating brain mechanisms involved. The purpose of this study examined how readers made these inferences by using Near-infrared spectroscopy (NIRS) which monitored brain activities during reading. It’s a non-invasive technique tracking the optical signals resulting from changes in the concentration of oxyhemoglobin and deoxyhemoglobin.

Method
There were 20 college students read 36 experimental texts sentence by sentence in a self-paced manner. This study was a 2 (overt versus zero pronoun) x 2 (strong versus weak causality) x 8 (brain areas) design. Each text consisted of two sentences, the first one describing two characters engaging in an activity, and the second one was a consequence or another situation. The degree of causality and the presence/ absence of pronoun were manipulated between sentences. A comprehension question was administered to probe their accuracy. The probe location was placed on the right and left hemispheres (inferior frontal gyrus, middle frontal gyrus, inferior parietal lobule, and middle temporal gyrus).

Results
For high causal conditions, the oxyhemoglobin is larger for sentences with pronoun than zero pronoun in the inferior and middle frontal gyrus. However, there are no differences for low casual conditions. For conditions with pronoun, the oxyhemoglobin is larger for high than low causal sentences¸ but there are no differences for zero pronoun conditions.

Conclusions
The patterns in this study indicate the importance of causality and pronoun for the online inferences in Chinese. Our results also imply casual and pronominal inferences activate different brain areas.